OBJECTIVE: To develop an original-mirror alignment associated deep learning algorithm for intelligent registration of three-dimensional maxillofacial point cloud data, by utilizing a dynamic graph-based registration network model (maxillofacial dynamic graph registration network, MDGR-Net), and to provide a valuable reference for digital design and analysis in clinical dental applications. METHODS: Four hundred clinical patients without significant deformities were recruited from Peking University School of Stomatology from October 2018 to October 2022. Through data augmentation, a total of 2 000 three-dimensional maxillofacial datasets were generated for training and testing the MDGR-Net algorithm. These were divided into a training set (1 400 cases), a validation set (200 cases), and an internal test set (200 cases). The MDGR-Net model constructed feature vectors for key points in both original and mirror point clouds (X, Y), established correspondences between key points in the X and Y point clouds based on these feature vectors, and calculated rotation and translation matrices using singular value decomposition (SVD). Utilizing the MDGR-Net model, intelligent registration of the original and mirror point clouds were achieved, resulting in a combined point cloud. The principal component analysis (PCA) algorithm was applied to this combined point cloud to obtain the symmetry reference plane associated with the MDGR-Net methodology. Model evaluation for the translation and rotation matrices on the test set was performed using the coefficient of determination (R²). Angle error evaluations for the three-dimensional maxillofacial symmetry reference planes were constructed using the MDGR-Net-associated method and the "ground truth" iterative closest point (ICP)-associated method were conducted on 200 cases in the internal test set and 40 cases in an external test set. RESULTS: Based on testing with the three-dimensional maxillofacial data from the 200-case internal test set, the MDGR-Net model achieved an R² value of 0.91 for the rotation matrix and 0.98 for the translation matrix. The average angle error on the internal and external test sets were 0.84°±0.55° and 0.58°±0.43°, respectively. The construction of the three-dimensional maxillofacial symmetry reference plane for 40 clinical cases took only 3 seconds, with the model performing optimally in the patients with skeletal Class Ⅲ malocclusion, high angle cases, and Angle Class Ⅲ orthodontic patients. CONCLUSION This study proposed the MDGR-Net association method based on intelligent point cloud registration as a novel solution for constructing three-dimensional maxillofacial symmetry reference planes in clinical dental applications, which can significantly enhance diagnostic and therapeutic efficiency and outcomes, while reduce expert dependence.
Humans ; Deep Learning ; Algorithms ; Imaging, Three-Dimensional/methods* ; Male ; Female ; Maxilla/diagnostic imaging* ; Adult

Objective:To establish an intelligent registration algorithm under the framework of original-mirror alignment algorithm to construct three-dimensional (3D) facial midsagittal plane automatically. Dynamic Graph Registration Network (DGRNet) was established to realize the intelligent registration, in order to provide a reference for clinical digital design and analysis.Methods:Two hundred clinical patients without significant facial deformities were collected from October 2020 to October 2022 at Peking University School and Hospital of Stomatology. The DGRNet consists of constructing the feature vectors of key points in point original and mirror point clouds (X, Y), obtaining the correspondence of key points, and calculating the rotation and translation by singular value decomposition. Original and mirror point clouds were registrated and united. The principal component analysis (PCA) algorithm was used to obtain the DGRNet alignment midsagittal plane. The model was evaluated based on the coefficient of determination (R 2) index for the translation and rotation matrix of test set. The angle error was evaluated on the 3D facial midsagittal plane constructed by the DGRNet alignment midsagittal plane and the iterative closet point (ICP) alignment midsagittal plane for 50 cases of clinical facial data. Results:The average angle error of the DGRNet alignment midsagittal plane and ICP alignment midsagittal plane was 1.05°±0.56°, and the minimum angle error was only 0.13°. The successful detection rate was 78% (39/50) within 1.50° and 90% (45/50) within 2.00°.Conclusions:This study proposes a new solution for the construction of 3D facial midsagittal plane based on the DGRNet alignment method with intelligent registration, which can improve the efficiency and effectiveness of treatment to some extent.

Objective To determine whether prenatal ultrasonography (US) score is more effective than renal pelvic anterior posterior diameter (PAPD) for the prognostic evaluation of fetal hydronephrosis.Methods Fetuses with hydronephrosis (PAPD≥ 10 mm) were examined by prenatal US in the third trimester.PAPD,renal parenchyma thickness (RPT) and pelvicaliceal morphology (PM) were measured and graded from 0 to 3 score on the basis of severity of hydronephrosis,then the total US score of each kidney was obtained.According to the follow-up results after birth,all the cases were divided into two groups:physiological and pathological hydronephrosis.Via Z test,paired comparison was made to analyze area under the curve (AUC) of US score and each of the other three factors.Results Confirmed by postnatal US and other clinical examinations,of 198 kidneys (158 cases ) with hydronephrosis,139 (70.20％ ) were physiological hydronephrosis and 59 (29.80％ ) were pathological hydronephrosis.AUC of PAPD,RPT,PM,US score was 0.897 (minimum),0.957,0.944 and 0.982 (maximum) respectively,and there was significant difference between AUC of US score and each of the other three ( P ＜0.05).US score was the best approach for differential diagnosis of fetal hydronephrosis.Conclusions Prenatal US score is more effective and accurate than the single factor (PAPD,RPT,PM) to differentiate fetal physiological and pathological hydronephrosis.It was a new quantitative method to evaluate the prognosis of fetal hydronephrosis,and should be disseminated and applied clinically.

Objective To evaluate the hemodynamic features of VX_2 tumor and peri-neoplastic liver parenchyma with low mechanical index gray-scale contrast enhancement.Methods Ultrasound contrast agent SonoVue((0.1) ml/kg) was applied respectively in 8 VX_2-bearing rabbits by intravenous bolus injection.Corresponding parameters of the time-intensity curve:time to enhancement(ET),time to peak intensity(PIT),peak signal intensity(PSI) and mean transit time(MTT) were measured using low mechanical index contrast gray-scale imaging and Wash-in/Wash-out time-intensity curve software.Results Gray-scale imaging delineated clearly the dynamic enhancement of the VX_2 tumor and the surrounding liver parenchyma.The ET and PIT were definitely earlier,the PSI lower and the MTT absolutely shorter in VX_2 tumors than those in the liver parenchyma.There were significant differences of parameters derived from the time-intensity curve between VX_2 tumors and liver parenchyma(P

Objective To evaluate the performance of artificial neural network(ANN) models for predicting ovarian malignancy in the patients with adnexal masses by using B-mode and color Doppler ultrasonography.Methods The data of 180 patients with adnexal masses were randomly divided into training and testing subsets.The training subsets were used to screen out significant single parameters and to compute the optimum statistic equations and to train the ANN.The testing subsets were used to estimate the performance of each of the two models in predicting ovarian malignancy.Results Thirteen of seventeen parameters were significant to diagnose malignancy in adnexal masses by single analysis(P

Objective To evaluate the role of interventional ultrasound technique in prenatal diagnosis of fetal chromosomal abnormalities.Methods Percutaneous ultrasound-monitored amniocentesis and cordocentesis were performed on 186 pregnant women with indicantions for prenatal diagnosis to detect karyotype of the fetus.Results All the cases were punctured successfully.Percutaneous ultrasound-monitored amniocentesis was performed in 105 pregnant women.The archievement rate of the aminiotic fluid was 97.1%,and the detection rate of the chromosomal abnormality was 8.8%.Percutaneous ultrasound-monitored cordocentesis were performed in 81 pregnant women.The archievement rate of the cord blood was 98.8%,and the detection rate of the chromosomal abnormality was 5.0%.There was no significant deviation between them.The valid indications of the puncture included the abnormal serology screening results of pregnant women,the history of abnormal deliveries,the history of the trisomy 21,and the abnormal fetus detected by ultrasound,the last of which was higher than the other indications at the detection of the chromosomal abnormality.Conclusions Interventional ultrasound technique proves to be valid in the prenatal diagnosis,and ultrasound detection of abnormal fetus indicates the possibility of the fetal chromosomal abnormality.

Objective To evaluate the diagnostic value of three dimensional transvaginal sonograpy( 3D TVS) for uterine cavity lesions.Methods Two dimensional transvaginal sonography(2D TVS) and 3D TVS were performed on 46 patients with irregular vaginal bleeding.The diagnosis were compared with the results of pathology and hysteroscopy.Results Of 46 patients,16 cases were diagnosed as endometrial hyperplasia,12 cases as endometrial polyp,9 cases as submucous myoma,3 cases as endometrial cancer, and the remaining 6 cases as normal secretive phase.To endometrial hyperplasia,the specificity,sensitivity,accuracy of 2D TVS and 3D TVS was 66.7% , 93.7% , 76.1% and 90.0% ,100%, 93.5% respectively.To endometrial polyp,the specificity,sensitivity,accuracy of 2D TVS and 3D TVS was 88.2% , 58.3% , 80.4% and 100%, 83.3% , 95.7% respectively.Conclusions 3D TVS improves the diagnosis of uterine cavity lesions,3D TVS combined with 2D TVS is the first choice for uterine cavity lesions.

Objective To study the feasibility of ultrasound-guided interventional therapy for parovarian cysts. Methods Diagnostic characteristics of 82 patients preoperatively diagnosed as having parovarian cysts were analyzed, and their clinical parameters were compared between 27 patients receiving ultrasound-guided interventional therapy and 55 patients receiving open or laparoscopic operations. Results Ultrasound examination was the principal means of diagnosis of parovarian cysts. The diagnostic accordance rate was 95.3% (82/86). The operative time, intraoperative blood loss and hospitalization costs in patients receiving ultrasound-guided intervention significantly decreased as compared with those receiving open or laparoscopic operation ( t =-20 798, 65 356, -184 533 , respectively; P =0 000). There were no significant differences in the postoperative recurrence rate and pyrexia rate between the two groups ( ? 2 =0 000, P =1 000; ? 2 =0 373, P =1 000). Conclusions Ultrasound-guided interventional therapy for parovarian cysts is feasible and offers satisfactory effects.

Objective To study the in vitro effects of sonodynamic management on human ovarian cancer cell line HRA. Methods Hematoporphyrin was selected as a sonosensitizer and ultrasound with certain intensity was used to activate hematoporphyrin.Then MTT assay and clony-forming efficiency assay were applied to determine the growth inhibitory effects of ovarian cancer cells. Electron microscope was applied to detect the morphological changes of the cells. Cell apoptosis was analyzed using flow cytometry. Results Hematoporphyrin had no significant inhibitory effects on the cells. However, when hematoporphyrin was used before ultrasound,it could enhance the killing effects of ultrasound(P

Objective To determine the transfection of PTEN gene in human ovarian carcinoma SKOV3 cells by ultrasound mediating and contrast agent′s enhancing, and to explore the mechanism of PTEN gene inhibiting the tumor cells′ invasion. Methods After addition of PTEN gene and 5 ?l lipofectamine,human ovarian carcinoma SKOV3 cells on 6-well plates were explored to ultrasound for 60 s in the presence(200 ?l) of ultrasound agent. The invasiveness of transfected cells was measured quantitatively by Matrigel invasion assays(Transwell chamber) after 48 h. Results The invasiveness of SKOV3 cells transfected by ultrasound-mediated PTEN gene was significantly declined compared with that of the controls. Conclusions The transfection of PTEN gene into SKOV3 cell can inhibit their invasion.The gene transfection can be enhanced by ultrasound and contrast agent,which may be expected to be a new means of clinical gene therapy.